Architect the Universal Agent: Design the central orchestration layer for our "Elevate" platform, allowing the AI to autonomously perform complex actions across integrated marketing tools.
Implement MCP Strategy: Lead the adoption of the Model Context Protocol (MCP) to standardize how our agents connect to external APIs, databases, and third-party platforms.
State & Memory Management: Architect the system’s persistent memory using Vector Databases (RAG) to ensure agents maintain context across long-term interactions and distinct sessions.
Client & Agency Automation
Bespoke Agent Development: Prototype and deploy specialized AI agents for our B2B clients (e.g., Lead Qualification Bots, Data Analysis Pipelines, Automated Reporting).
Operational Efficiency: Analyze internal agency processes and replace manual workflows with robust, error-tolerant automation.
Stack Selection: Assess specific project requirements to determine the optimal approach—choosing between low-code orchestration (n8n) or custom code (Python/TypeScript) based on scalability and complexity needs.
Technical Implementation
Advanced Integrations: Build custom connectors for undocumented or complex APIs, handling authentication (OAuth2, API Keys) and data transformation programmatically.
Custom Tooling: Extend the capabilities of standard automation platforms by writing custom nodes, scripts, and microservices in TypeScript or Python.
Reliability Engineering: Ensure all workflows are production-grade, implementing comprehensive error handling, logging, and rate-limit management strategies.
WHAT YOU BRING:
Engineering Background: 5+ years in backend-focused software development with a strong portfolio of AI/automation projects.
Strong Coding Skills: Production-level TypeScript (Node.js) or Python for extending low-code systems when needed.
AI/LLM Architecture: Deep understanding of function calling/tool use, embeddings, RAG, and prompt engineering (OpenAI, Anthropic).
API Architecture: Experience with REST, GraphQL, webhooks, and reverse-engineering private APIs.
Data & Storage: Proficiency with SQL (PostgreSQL) and vector stores (Pinecone, Weaviate, Supabase).
Product Mindset: Ability to work on a long-term AI product vision and influence the technical stack and architecture.
Nice to have:
WHAT WE OFFER:
USD compensation that values your expertise.
Work from anywhere: Fully remote to suit your lifestyle.
31 days of paid time off: 21 days of annual leave + 10 days of sick leave, because your health and work-life balance matter.
Growth-focused environment: Access to learning resources and clear pathways for growth.
Fun team events: Virtual cooking classes, yoga sessions, team quizzes and more
A culture of trust: We cut the red tape — results over rules always. Open talk, ownership, and getting things done together.
STEPS OF THE SELECTION PROCESS:
Screening Interview (45 minutes) – Initial conversation with a recruiter.
Test Assignment (2 days) – Complete a task to showcase your skills.
Team Interview (up to 60 minutes) – A meeting with your potential Team Lead and future colleagues to assess team fit and discuss role-specific responsibilities, expectations and workflows.
Technical Interview (up to 60 minutes) – A deep dive into your functional expertise, problem-solving approach, and practical skills relevant to the role.
Offer – If successful, you’ll receive an offer to join our team.
AI Digital does not discriminate on the basis of race, color, religion, gender identity or expression, national origin, age, military service eligibility, veteran status, sexual orientation, marital status, disability, or any other protected class in terms of employment. We support workplace diversity and believe strongly that it contributes to a broader collective perspective that consistently leads to better products and a better company. We are working hard to increase the diversity of our team.
Note: This role is open to applicants from:
EMEA (selected countries)
Please note that we’re unable to consider applications from outside these regions. Thanks for your understanding!
